Photothermal Radiometry Data Analysis by Using Machine Learning.

Photothermal techniques are infrared remote sensing techniques that have been used for biomedical applications, as well as industrial non-destructive testing (NDT). Machine learning is a branch of artificial intelligence, which includes a set of algorithms for learning from past data and analyzing new data, without being explicitly programmed to do so. In this paper, we first review the latest development of machine learning and its applications in photothermal techniques. Next, we present our latest work on machine learning for data analysis in opto-thermal transient emission radiometry (OTTER), which is a type of photothermal technique that has been extensively used in skin hydration, skin hydration depth profiles, skin pigments, as well as topically applied substances and skin penetration measurements. We have investigated different algorithms, such as random forest regression, gradient boosting regression, support vector machine (SVM) regression, and partial least squares regression, as well as deep learning neural network regression. We first introduce the theoretical background, then illustrate its applications with experimental results.

The Effect of Sun Tan Lotion on Skin by Using Skin TEWL and Skin Water Content Measurements.

Stratum corneum (SC) is the outermost skin layer. SC hydration is important for its cosmetic properties and barrier function. SC trans-epidermal water loss (TEWL) measurements and skin water content measurements are two key indexes used for SC characterisation. The instrument stability and accuracy are vitally important when measuring small changes. In this paper, we present our latest study on the effect of sun tan lotion on skin by using skin TEWL and skin water content measurements. We developed techniques to improve the measurement stability and to visualise small changes, as well as developed machine learning algorithms for processing the skin capacitive images. The overall results show that TEWL and skin water content measurements are capable of measuring the subtle changes of skin conditions due to the application of sun tan lotions. The results show that the TEWL values decreased after the sun tan lotion application. The sun tan lotion with SPF 20 had the lowest decrease, whilst the sun tan lotion with SPF 50+ had the highest decrease. The results also show that the skin water content increased after the sun tan lotion application, with SPF 20 having the highest increase, whilst SPF 50+ had the lowest increase.

An Effective Conversion of Visemes to Words for High-Performance Automatic Lipreading.

As an alternative approach, viseme-based lipreading systems have demonstrated promising performance results in decoding videos of people uttering entire sentences. However, the overall performance of such systems has been significantly affected by the efficiency of the conversion of visemes to words during the lipreading process. As shown in the literature, the issue has become a bottleneck of such systems where the system's performance can decrease dramatically from a high classification accuracy of visemes (e.g., over 90%) to a comparatively very low classification accuracy of words (e.g., only just over 60%). The underlying cause of this phenomenon is that roughly half of the words in the English language are homophemes, i.e., a set of visemes can map to multiple words, e.g., "time" and "some". In this paper, aiming to tackle this issue, a deep learning network model with an Attention based Gated Recurrent Unit is proposed for efficient viseme-to-word conversion and compared against three other approaches. The proposed approach features strong robustness, high efficiency, and short execution time. The approach has been verified with analysis and practical experiments of predicting sentences from benchmark LRS2 and LRS3 datasets. The main contributions of the paper are as follows: (1) A model is developed, which is effective in converting visemes to words, discriminating between homopheme words, and is robust to incorrectly classified visemes; (2) the model proposed uses a few parameters and, therefore, little overhead and time are required to train and execute; and (3) an improved performance in predicting spoken sentences from the LRS2 dataset with an attained word accuracy rate of 79.6%-an improvement of 15.0% compared with the state-of-the-art approaches.

Micro-relief analysis with skin capacitive imaging.

BACKGROUND: In this study, the performance of capacitive imagining in skin micro-relief analysis was investigated. This measurement principle has been used for skin hydration measurements over the last decade and it is commercially available by various manufacturers. Strengthening its potential for new applications could offer an affordable and portable multi-purpose device for in vivo skin research. Previous studies in the literature have used a wide range of optical devices to determine how the skin surface topographic features are affected by chronological age, environmental influences and living habits. MATERIAL AND METHODS: A capacitive system was used in order to capture hydration images from the middle volar forearm of twelve volunteers. The visual output of the system was studied and image processing algorithms were adapted to automatically extract skin micro-relief features. The change in the skin network of lines during arm extension, the lines' anisotropy index and the number of closed polygons per skin surface area were plotted against the subjects' chronological age. The results were compared with optical measurements from the literature to validate our algorithms and evaluate the capacitive imaging in skin micro-relief analysis. RESULTS: The change in the intensity of primary and secondary lines during arm extension and the number of closed polygons per surface area were in agreement with the literature. The anisotropy index output gave inconclusive results. CONCLUSIONS: The experimental results show that the capacitive systems could only extract two-dimensional skin topographic features.

Skin integrity testing and monitoring of in vitro tape stripping by capacitance-based sensor imaging.

BACKGROUND: Despite the frequent use of porcine ear skin for tape-stripping experiments, the peculiarities of this skin type have not been characterised in detail yet. Thus, different techniques were employed to investigate the skin surface structure of porcine ear skin and the changes in barrier function during in vitro tape stripping. To this end, the potential of capacitance-based skin hydration imaging as a means of skin quality control was investigated for the first time. METHODS: The porcine ear model was characterised before and during tape stripping using transepidermal water loss (TEWL) measurements, capacitance-based sensor imaging, attenuated total reflection Fourier transform infrared spectroscopy (ATR-FTIR) and optical light microscopy. RESULTS: The capacitance-based sensor was found to deliver precise information about the quality of the employed skin sites before and during tape stripping. The removal of stratum corneum proteins was highly reproducible even for different porcine ear types. The mean greyscale values showed an excellent linear correlation to the corresponding TEWL values and the respective penetration depth. Optical light microscopy confirmed the presence of canyons on the surface of porcine ear skin. CONCLUSION: The results suggest that the capacitance-based sensor is a suitable tool for skin integrity testing of porcine ear skin in vitro and for monitoring changes in skin barrier function.

In vivo skin capacitive imaging analysis by using grey level co-occurrence matrix (GLCM).

We present our latest work on in vivo skin capacitive imaging analysis by using grey level co-occurrence matrix (GLCM). The in vivo skin capacitive images were taken by a capacitance based fingerprint sensor, the skin capacitive images were then analysed by GLCM. Four different GLCM feature vectors, angular second moment (ASM), entropy (ENT), contrast (CON) and correlation (COR), are selected to describe the skin texture. The results show that angular second moment increases as age increases, and entropy decreases as age increases. The results also suggest that the angular second moment values and the entropy values reflect more about the skin texture, whilst the contrast values and the correlation values reflect more about the topically applied solvents. The overall results shows that the GLCM is an effective way to extract and analyse the skin texture information, which can potentially be a valuable reference for evaluating effects of medical and cosmetic treatments.

Two dimensional finite element modelling for dynamic water diffusion through stratum corneum.

Solvents penetration through in vivo human stratum corneum (SC) has always been an interesting research area for trans-dermal drug delivery studies, and the importance of intercellular routes (diffuse in between corneocytes) and transcellular routes (diffuse through corneocytes) during diffusion is often debatable. In this paper, we have developed a two dimensional finite element model to simulate the dynamic water diffusion through the SC. It is based on the brick-and-mortar model, with brick represents corneocytes and mortar represents lipids, respectively. It simulates the dynamic water diffusion process through the SC from pre-defined initial conditions and boundary conditions. Although the simulation is based on water diffusions, the principles can also be applied to the diffusions of other topical applied substances. The simulation results show that both intercellular routes and transcellular routes are important for water diffusion. Although intercellular routes have higher flux rates, most of the water still diffuse through transcellular routes because of the high cross area ratio of corneocytes and lipids. The diffusion water flux, or trans-epidermal water loss (TEWL), is reversely proportional to corneocyte size, i.e. the larger the corneocyte size, the lower the TEWL, and vice versa. There is also an effect of the SC thickness, external air conditions and diffusion coefficients on the water diffusion through SC on the resulting TEWL.